INDEX

Explanations

names or terms related to legal cases and law enforcement

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

«ĺ

-0.75

ĺħ

-0.74

 mileage

-0.72

ģĸ

-0.70

 anecd

-0.68

 impulse

-0.67

 uranium

-0.65

Ĥª

-0.65

 inference

-0.63

ESC

-0.63

POSITIVE LOGITS

pering

1.33

pered

1.31

borgh

1.01

eless

0.98

iami

0.98

Tam

0.96

arin

0.95

pling

0.95

my

0.93

riel

0.92

Activations Density 0.019%

names or terms related to legal cases and law enforcement

No Comments

No Known Activations